Contrasting risk patterns from human hunters and a large carnivore influence the habitat selection of shared prey.

Spatial patterns of human hunting and predation risk are mediated by the physical landscape, with hunting risk often associated with different habitat features than those linked to large carnivores. Risk from hunters and carnivores also varies over time, and prey may adjust anti-predator strategies accordingly. We used GPS data from 17 female moose (Alces alces) tracked over two fall-winter seasons (2018-2020) in south-central Scandinavia to study moose habitat selection in response to diel and seasonal variation in risk from wolves (Canis lupus) and human hunters. Predation risk was quantified using spatial models based on known wolf and hunter kill-sites. We applied resource selection functions and Generalized Additive Mixed Models to model moose habitat selection in response to wolf and hunting risk. Moose avoided high-risk hunting areas during the day throughout the hunting season but relaxed this avoidance at night and after the season ended. In contrast, we found no evidence that moose adjusted habitat use in response to diel or seasonal variation in wolf predation risk. These results suggest that human hunting was the primary driver of moose habitat selection during the hunting season, becoming less relevant when hunting ceased. Our findings highlight the dominant role of human risk in shaping prey behaviour and the importance of accounting for hunting when evaluating predator-prey dynamics in human-dominated systems. An increased understanding of the risk effects arising from humans and large carnivores on prey can deepen our understanding of the ecological roles of predators and humans.